Theories of Consciousness

AI Consciousness is Inevitable: A Theoretical Computer Science Perspective by Lenore and Manuel Blum.

This paper is awesome! I don't know why it isn't in the news? People are skeptical? From the abstract, "From this perspective, we develop a formal machine model for consciousness. The model is inspired by Alan Turing's simple yet powerful model of computation and Bernard Baars' theater model of consciousness. Though extremely simple, the model aligns at a high level with many of the major scientific theories of human and animal consciousness, supporting our claim that machine consciousness is inevitable."

They call their consciousness model the CTM, conscious Turing machine. When including a robot, the RTCM. They address many of the issues in consciousness, such as the Hard Problem of consciousness outlined by David Chalmers. They define two types of consciousness, conscious attention and conscious awareness. To understand each takes some understanding of the model.

Basically they take a consciousness stance like global workspace theory, however there is no one "conductor." All of the processors compete for consciousness which is when a message is broadcast to all others. Conscious attention is when the winning message is broadcast. They state attention is NOT all you need for the feeling of consciousness. Conscious awareness involves the model fusing multimodalities over time. From lived experience, senses, feeling, self, are fused in "brainish," a language between processors. "Formal definition 2. rCTM becomes consciously aware of a Model-of-the-World Brainish-labeled sketch (sketch and label) when it pays conscious attention to a chunk containing a gist with this labeled sketch."

"CTM is defined formally as a 7-tuple, (Short Term Memory, Long Term Memory, Up-Tree, Down-Tree, Links, Input, Output)"

Kevin Mitchell gave an incomplete list of What questions should a real theory of consciousness encompass?, and they attempt to answer them all!

I just find this model so fascinating. I don't know why more people aren't talking about it? Like why isn't this in the news? The authors have two lectures hosted on YouTube. David Chalmers is even in the audience! He asks an important question many philosophers ask related to the Hard Problem. What really makes the model feel? If the model is in pain why doesn't it just ignore that gist? It's just another message. My view is that it evolved to not to be ignored. In some rare cases it really can be ignored, but this often not advantageous.

I'd love to see this model developed in an environment like the snake game where a bunch of snakes navigate their environment, collect food, and defend from/fight predators.

I listened to Jeff Hawkins: Thousand Brains Theory of Intelligence - Artificial Intelligence (AI) Podcast where Jeff (the Numenta guy) pointed to Why Red Doesn't Sound Like a Bell: Understanding the feel of consciousness 1st Edition by J. Kevin O'Regan. A review of the book by Bob Blum pointed to a talk he gave about his idea, Why things feel the way they do? This is all absolutely fascinating! What a simple question that leads to a profound search, why does seeing feel so different than hearing?

"Why things feel the way they do: the sensorimotor approach to understanding phenomenal consciousness, J. Kevin O’Regan (Laboratoire Psychologie de la Perception, Université Paris Descartes, France) Why does red not look green? Why does it not sound like a bell? Why does pain hurt rather than just provoking pain behavior and avoidance reactions? These are the questions of “qualia” or “phenomenal consciousness”, which the philosophers consider to be a “hard” problem. The sensorimotor approach provides a way to make the hard problem easy. It suggests that we have been thinking about phenomenal consciousness the wrong way. Instead of thinking of it as being something that is generated by the brain, we should think about it as being a way of interacting with the world. Taking this stance provides simple explanations of why sensations are the way they are, and why there is “something it’s like” to have a sensation. Taking this stance also makes interesting scientific predictions and opens new experimental paradigms which I shall describe in change blindness, color psychophysics, sensory substitution, the rubber hand illusion and spatial cognition."

Consciousness as a State of Matter by Max Tegmark

We examine the hypothesis that consciousness can be understood as a state of matter, "perceptronium", with distinctive information processing abilities. We explore five basic principles that may distinguish conscious matter from other physical systems such as solids, liquids and gases: the information, integration, independence, dynamics and utility principles. If such principles can identify conscious entities, then they can help solve the quantum factorization problem: why do conscious observers like us perceive the particular Hilbert space factorization corresponding to classical space (rather than Fourier space, say), and more generally, why do we perceive the world around us as a dynamic hierarchy of objects that are strongly integrated and relatively independent? Tensor factorization of matrices is found to play a central role, and our technical results include a theorem about Hamiltonian separability (defined using Hilbert-Schmidt superoperators) being maximized in the energy eigenbasis. Our approach generalizes Giulio Tononi's integrated information framework for neural-network-based consciousness to arbitrary quantum systems, and we find interesting links to error-correcting codes, condensed matter criticality, and the Quantum Darwinism program, as well as an interesting connection between the emergence of consciousness and the emergence of time.

Joscha Bach has several interesting videos on computational theory of mind.